Why I'm Staying Out Of The Substack Religion Debate
Read the original on Astral Codex Ten →
Summary
Scott explains why the 2020s Substack revival of the God debate (Adelstein/Bentham's Bulldog vs Chappell and others) leaves him cold: his atheism precedes the object-level arguments. Part I uses an 'aliens caused Napoleon' analogy. Any anomaly can be explained by a powerful hidden entity that intended that exact outcome, so such explanations are maximally overfit and need a pre-emptive penalty. Conspiracy theories are the weak case; gods (hidden, exotic substrate, goals only loosely matching the outcome) are the strongest case, and the history of storm and sun gods shows the pattern retreating as anomalies were explained. Part II rebuts the divine-simplicity reply ('God is mind without limits, hence simplest'): subtracting limits is subtracting negatives, i.e. adding capability. A reductio follows: factories would import aircraft-without-limits and laboriously add constraints to get biplanes. Part III anticipates the dualist escape (Mind as a real unbounded kind). He separates consciousness, which may be mysterious, from computation, which demonstrably runs on brains: lesions, drugs, genes, sleep, understood circuits, and AI neural nets as direct proof. Unbounded computation therefore can't ride on an exotic substrate, and theism has to carry the full overfitting penalty. Closes with a self-aware joke about participating by declaring non-participation.
Why this score
Quality 69 · Strong. Strong, upper-middle. A tight, well-built essay with two memorable devices: the alien-Napoleon overfitting frame and the 'subtracting negatives' takedown of divine simplicity with the biplane-factory reductio. Part III's unbundling of consciousness from computation is the most useful philosophical move. It is still a short response piece in a well-trodden debate that relies on intuition pumps rather than a formal treatment (he admits the Kolmogorov version is beyond most readers). Above the Tegmark proofs-of-God post (62), roughly level with Against The Cultural Christianity Argument (73); not an Excellent-band reframe.
Claude’s paradigm shift 48 · Moderate. Moderate. Overfitting and complexity penalties against God-hypotheses have a long lineage: Hume, Dawkins's 'Ultimate Boeing 747', Yudkowsky on Occam and Kolmogorov. The fresh parts are the 'subtracting limits is adding capability' formulation aimed at the Swinburne/Adelstein simplicity argument, and making the computation-vs-consciousness unbundling the hinge of the dualist reply. A sharpening, not a new frame.
Real-world impact 1 · Negligible. 1: an intervention in a live Substack debate. Too recent to show measurable uptake beyond the blogosphere exchange it joins.